Can you articulate "what is wrong with the communities" discussed here? Personally, I haven't run into anything particularly "wrong" with the EN WIKT community. It isn't huge, sure, but then it never has been, at least for the years I've been involved (some activity in the first half of 2006, then a hiatus until end-2010, participating off-and-on since then).
If what is wrong is not clearly defined, we have no objective measurement for discussing how effective Flow is at "fixing" anything. All we have is our subjective reactions to our experiences using Flow.
About "shrinking" communities, I myself have participated much less in the EN Wikipedia over the years, in large part because it's mostly done -- any of the articles that I might think of to write, have already been written. Maintenance is much less interesting, and requires fewer people. A shrinking editor base at this point is not a problem to solve, and is instead both expected and inevitable. This is part of the natural progression of any encyclopedia project: once the articles are written, you don't need lots of editors anymore, beyond a relatively smaller group to maintain and update existing material. This latter group also often requires a different set of skills and interests than the group who writes the articles initially.
I fear that Flow is a response to the latter issue (i.e. developers have grown bored with maintenance and are coming up with solutions in search of problems).
Back on topic, your post suggests that one of the things you think is "wrong" is that we have a shrinking editor base, and that "we need new people in the project". Can you explain why? And can you explain any other issues that are "wrong", that Flow might somehow fix?