i think solo/small team hacks (which have a better track record regarding completion and tend to have a stronger identity than huge community projects) and writing guides and game documentation (which make modding the game more accessible) are more reliable methods of boosting scene activity. from what ive gleaned community remix was incredibly lofty and lacked focus, and died despite being conceptualized back when the scene was (to my understanding) newer and more active. additionally, participation in other community events (like level contests) has been limited; ive also seen people having trouble recruiting people to help with their own projects. the only successful community project im aware of is the 3 years of nsmbhd hack— which while ultimately greater than the sum of its parts, it doesnt seem to have much identity beyond "a nsmb hack made with community contributions", and had a more limited scope compared to community remix.
practically speaking, in regards to increasing community activity/publicity, the output doesnt seem worth the input. logistically speaking, i dont think enough people have the time or resources needed to work on it.

The main problem with making a comunity project is that people dont end up making levels, we have tried holding level compititons on nsmb central but there really isnt an intrest in participating since most active nsmb hackers are to busy with their hacks.