There are constant references on forums to how they want to move away from XUL, for both Firefox and Thunderbird. XUL works, but I feel like very people are able to improve the guts of XUL, and that can be a huge risk for the future of the project. Pale Moon seems to be the main contender willing to maintain XUL, but presumably not improve it (changes to XUL can break extensions).
As someone working on another free software project with an aging code base, we see this debate all the time: limited funds, limited interest to maintain the old core, but users rely on it. However, users also want performance improvements, more reliability, new features... not to mention what do you do when security issues pop up. At some point you need to cut your losses and find a model easier to maintain.
