The main question, I think, is if the goal is worth the work. See, for a hook/scripting language/server tool/etc the work can improve jazz2 immensily (see the possibilities with Neobeo's hook, or how useful Project Controller is). The goal is worth the work in these cases: it pays off. A JCS clone would be a substantial amount of work, and would it pay off?

Consider that:
1a There seems to be a great lack of talented level/tileset creators (the human kind, not the program kind). This lack of people is greater than there is a lack of jcs tools.
1b Only a couple of people have taken JCS 'to the max' and exploit its current functionality.
1c This tool would therefore be mostly beneficial to the small elite of talented people, and not neccessarily beneficial to the greater public.
2 Recreating JCS + a tileset compiler is a huge amount of work
3 There are many things that would be (more?) useful to the community right now. A good chatlogger, scripting abilities, server tools (level voting, kickvoting, etc), security tools.
3 There aren't many good programmers in the community.
4 Their time may be better spent (it's their decision though, of course, and they're free to do whatever they want) on things that have a bigger work/payoff ratio.
5 If the additions would just be some textboxes/colour pickers/etc it would be odd to recreate an entire program for it
6 A tileset creator could be created independendly from JCS if it's needed

Neobeo mentioned my online JCS. what I do think is that maybe, if there'd be some collaborative JCS/a JCS with version control/online repository features, that might be worth the work.