On top of that the only way to control a complex remote application was something like X because having 15 80x25 terminals wasn't cutting it.
The web browser and markup, combined with the advancement in local processing/rendering ability completely destroyed one of the fundamental problems X was trying to solve. That it didn't kill X right then and there was, in my opinion, an artifact of the implementation as a glorified telnet session for the HTTP server.
The rendering model for most apps is totally different.
The "display side" of Web apps can do a lot of useful processing which cannot be done on the "display side" of X apps. JavaScript is not my favorite technology, and the performance is, at best, mediocre, but you can perform useful computations with it.
Also, well written Web apps can reduce network activity to a minimum.
Consider, for a moment, if ten thousand people were using HN via an X app. HN would need monster servers and monster bandwidth to handle all that load, even if the users were only scrolling through comments or spending minutes typing up replies.
During the time I've been typing this reply, the Web app side of HN is making no network calls and putting no demand on HN's servers at all.
I've seen people vaguely equate X apps and Web apps before, and I'm always bothered by it since they have such important fundamental differences that change everything.
I used to rather like X back in the early 90's but I think we've moved on a bit since. The introduction of hardware accelerated graphics and the decline of the old "network application" model have pretty much killed the model it was built on.
I'd like to see things such as Wayland succeed in the future.