I think below a certain-size user base it's hard to ensure the health of such topic feeds, especially if going for a varied palate of feeds--the result instead could be a high percentage of malnourished topic feeds, and these can do the opposite of drive further adoption by reinforcing a sense of relative emptiness.

I think the Nostr user base would need to grow much larger first, and for other reasons, of which I think there are many. (Generally I think Nostr is the architecture-type of the future.)

Once the Nostr user base hits a certain size, let's say 500k daily actives, then this topic-feed/new-user-attraction positive feedback loop starts to kick in, as the feeds will be much better nourished. So at that point it makes sense to start to promote topic feeds at large. But before that point I think it makes sense to drive sign-ups based on other already-salient aspects of the protocol or individual clients.

(Putting aside exceptions for feeds of certain niche topics already popular here, and bridged-feeds for what's hot on other protocols/networks which do have those larger user bases.)