The example of the Winchester Manuscript (the version of Le Morte d'Arthur found in 1934) is pretty interesting and will help answer the OP's questions.

First, let's talk about Le Morte d'Arthur. This is a pretty long book (the Penguin edition runs to two volumes and elicits groans from undergraduates) composed by Sir Thomas Malory around 1470. It brings together a lot of material from various earlier French and English sources about the legendary King Arthur and his knights. So it's sort of the first compilation and rationalization of the Arthurian Legendary Universe that had grown up in the preceding centuries. Malory had time to put it together, by the way, because he was imprisoned for a number of unchivalrous acts. Le Morte d'Arthur was one of the first English books printed by William Caxton (the pioneer of English printing). Caxton stated that he had organized Malory's text into books and chapters. Two copies (one complete) of Caxton's 1485 edition survive. The book was reprinted several times over the next few centuries. While the term is anachronistic, it's not too far-fetched to call Le Morte d'Arthur the first secular English-language best-seller. Anybody interested in medieval English literature or Arthuriana has probably read it, and any subsequent venture into these topics is shaped by it. I believe Le Morte d'Arthur has been continuously in print since 1816.

So what happened in 1934? Walter F. Oakeshott, the assistant librarian of Winchester College (not a university but an independent boarding school--think Hogwarts but less spellcasting) found a medieval manuscript that contained most of Le Morte d'Arthur, probably copied in the 1470s. (By the way, "manuscript" refers specifically to a document written by hand, so the printed 1485 version was by definition not a manuscript.) Because of what it was and where it was found, this became known as the Winchester Manuscript.

The way in which Oakeshott found the Winchester Manuscript helps us understand how an early copy of such a famous work lurked for so long. Oakeshott was actually trying to compile a list of medieval bindings. Naturally he wished to examine Winchester College's collection of medieval manuscripts, which were kept under lock and key, but he was disappointed to see almost immediately that none of them had medieval bindings: they had all been rebound later. Nonetheless, he looked though the books and noted that one, in a red sheepskin binding, contained tales of King Arthur's knights and was heavily rubricated (use of red ink, in this case to set out the proper names). A few weeks later, Oakeshott happened to be reading Le Morte d'Arthur (a work he did not know well) and noticed similarities to the manuscript he'd found. In fact, the Winchester Manuscript was Le Morte d'Arthur, albeit missing the first and last quires (groups of pages) and with numerous variations from Caxton's text. This gave scholars of the text plenty to talk about since they could now see what Caxton may have changed; however, it's generally agreed that the Winchester Manuscript was not written by Malory himself (who wouldn't have had the skillset to copy such a book) and thus varied from his original text, and that Caxton didn't prepare his edition solely from the Winchester Manuscript.

Just to complete the story, the Winchester Manuscript was rebound in 1947 as a result of increased interest, then was acquired by the British Library in 1976, which is also around the time a facsimile edition was created. Facsimiles and digitization are ways of making the manuscript available (not just the text) without subjecting the artifact itself to use and abuse.

But let's talk more about the Winchester Manuscript. How did it come to be in Winchester College's library without anyone knowing what it contained? The short answer is that we don't know. We can be certain of the following:

--Watermarks on the paper suggest the paper was manufactured in the 1470s. (Yes, the Winchester Manuscript is a manuscript on paper.)

--It was copied by at least two scribes. This kind of thing is determined by close analysis of the handwriting. Looking at language lets scholars also develop theories about where the scribes were from, since we tend to think their manner of speech would affect their writing, even when they were ostensibly just copying.

--Caxton actually had the Winchester Manuscript in his shop when he printed his 1485 edition, because imprints on the Winchester Manuscript show that pages from the printed edition were laid on top of it. But we strongly suspect he had other manuscripts because there are some sections missing from the Winchester Manuscript (apparently because of scribal inattention) that Caxton has.

--Caxton may still have had it in 1489 or later, because a fragment of an indulgence he printed in 1489 was used to repair a page, but it's also possible someone else did the repair.

--The next clue is that a "Richard Followel" wrote his name and a short poem in the margin of a page. This individual has been identified as living in Litchborough, of which a branch of the Malory family were lords, and dying in 1535. Thus, there is a theory that the Winchester Manuscript was a fancy copy of Malory's work given to the author's Litchborough cousins and lent to Caxton for an extended period to assist in creating the printed edition, then returned.

--But we still don't know how it got to Winchester College. It must have been there no later than 1839 because it's noted in a catalogue at that time, where it is identified as "no. 13." The red sheepskin binding in which Oakeshott found the Winchester Manscript dates to around 1800.

So the Winchester Manuscript at some point between 1535 and 1839 passed from private hands to Winchester College. It's a matter of conjecture when specific knowledge of its contents was lost. Certainly Caxton knew, and the book played some role in his creation of the 1485 printed edition. On the other hand, the existence of that edition may have made manuscript copies--of which others must have existed--less valuable and interesting for a period, since if what you actually wanted to do was read Le Morte d'Arthur, the printed version was better. At some point the first and last quires were lost, making the Winchester Manuscript even less useful and less easy to identify. When the Winchester Manuscript was noted in the catalogue, either its identify as a manscript copy of Malory's famous work was not known or not felt worthy of noting, so it fell to Oakeshott--who ostensibly was "judging books by their covers" and just happened to look within--to figure it out nearly a century later.

So now we can start answering the OP's questions.

Yes, documents that old are nowadays considered important historical documents--but at one time they were just old books and papers nobody wants or takes much notice of. So yes, it was quite easy for such documents to get lost. This is particularly true for the late middle ages (14th-15th centuries) when there was a pretty dense forest of documentation of which much survives. Note that apparently for much of its history, the Winchester Manuscript was either in a private library or in a secluded part of a school library where it probably had little practical use, and possibly no one cared all that much about it or what was in it. And, of course, even if you knew it was a manuscript version of a pretty famous work, before modern scholarship with its interest in redaction and dialect and variants, you might have just thought it was pretty useless.

I'll also add that changes in media technology lead to a lot of stuff getting thrown out. Once you have printed books, why bother with manuscripts? Let me explain this by analogy. Imagine it is 2002 and while going through family articles you come across a videotape that contains most of Top Gun recorded off a television broadcast in 1993, with commercials but missing the first few minutes (because your dad tuned in a few minutes late) and the very end (because the tape ran out), plus some scenes were deleted (for runtime and content). Would you keep it as a historical artifact or throw it out because the movie's available on DVD? But it turns out that was one of the last copies of Top Gun as shown on television at that time and, unknown to you, a scholar of American media in 2100 would be able to make an interesting argument about decency standards and views of the body had they been able to access that tape.

Anyway, really interesting question and I hope this helps!