Comparative analysis of mythologies is a cornerstone of good historical folklore research. There is nothing wrong with this. In the case of various Indo-European pantheons and the stories associated with them, it is possible to deduce a great deal about what prehistoric traditions may have existed. This has also occurred with non-Indo-European stories.

One should not become too excited by the term "prehistoric," however. While a few attempts have been made to push this sort of reconstructive work back to ten thousand years before present (or even more) most of this reconstructive work is restricted to a thousand or more years before the earliest written records.

The second part of your question is where the problem resides. Many people find reason to declare that they have "figured out" what a given story or motif was "based on." These declarations are almost always speculative and created out of thin air. They are simply based on someone's imagination and likely the ingestion of a great deal of coffee. The problem is that some jerk with a PhD says that "he's figured out" a well known myth, and he gets an article in the New York Times and everyone starts talking about it as though that is the new cutting edge thinking about that myth.

That process is bad history, bad folklore, and bad reasoning. I don't give a shit what anyone imagines to be the case (or how much coffee they have had). Is there proof or is this just speculation? Almost always, there is no proof beyond coincidence (often not very good coincidence), and it is all speculation.

A really great introduction to the history of thought in the way folklorists have tackled these sorts of questions is Alan Dundes, editor, International Folkloristics: Classic Contributions by the Founders of Folklore (1999). It's a great collection of essays, and the introductions by Dundes are great (sometimes better than the essays). He was a masterful writer.