Not hieroglyphics, but you may be interested in the deciphering of linear B, a Mycenaean script for which there was no equivalent of the Rosetta stone, and that was successfully deciphered in the 50s. Simon Singh has a chapter on this in the Code Book where he explains how it was done exactly.

I have a post about how scholars decipher extinct writing systems here: http://www.reddit.com/r/AskHistorians/comments/25bh2p/if_we_had_never_discovered_the_rosetta_stone/

Side note: Cryptographers actually have a surprisingly bad track record with deciphering lost writing systems, which if you think about it makes sense given that cryptography is about figuring out how to read things you're not supposed to be able to read and decipherments are about figuring out how to read things that you are supposed to be able to read. The case of linear B is a good example; it was deciphered through an enormous amount of gruntwork that relied on looking for sign combinations that were repeated often enough to suggest that they were grammatical particles, recognizing place names and ethnic names in the script by working out which signs were likely to be vowel or initial-vowel, and deducing based on this evidence that the script was likely an archaic dialect of greek. The key tools of decipherment seem to be art history(at least in the case of Mayan), linguistics, some background on the likely historical contexts of the texts in question, good drawings of the texts being deciphered and an enormous amount of patience.

This post makes me curious to ask, how much of known 'religious' Egyptian hieroglyphs and 'normal' Egyptian has been deciphered to the best of modern linguistics ability?