Netscape was losing marketshare like crazy when the code was released. It had little to do with the long release time for Mozilla 1.0. Also, the code that Netscape released was unbuildable. They basically ripped out anything that had a copyright question on it so all you got an incoherent blob of C code that nobody understood. So the long release time had was mostly a result of trying to figure out what they had and how they'd go about turning it into a functioning computer program.
2) I've poked around in the Mozilla codebase and there is a TON of "mcom" stuff in there. So if there was a serious effort to rewrite it (I don't think there was) they sure did leave a lot of truly ancient code.