I don’t know for sure, but your “weekly ‘feeder’ or ‘home’ lists” may actually be more valuable for eBird than your “more unusual sightings” because regular complete lists are important for estimating species departures and arrivals, abundance, etc. But to be fair they are less enjoyable lists to make than posting unusual species, and the unusual sightings would be useful for local researchers or other interested people to learn about what’s in the area and how that changes over the years though.
Here is eBird’s official research page: https://ebird.org/science
I think their mapping projects require the complete presence and absence data that eBird checklists provide and iNaturalist observations don’t.
They also have this list of publications, but I don’t know if all of them used exclusively eBird data or if they pulled bird data from multiple sources through GBIF.
(Also see related thread here: https://forum.inaturalist.org/t/submit-the-same-sightings-to-both-ebird-and-inaturalist/1852)
I think this is a good point, the lack of any communication within eBird (so they don’t have to moderate) is a drawback. Although I sort of doubt that the birding community will move to iNaturalist, because it is pretty established on facebook with a large network of facebook groups. But perhaps changes to facebook will change that.
Here are a couple examples of discussion of significant rare birds originally found on iNaturalist: https://www.inaturalist.org/observations/2343708, https://www.inaturalist.org/observations/16295265
I’m sure there was also extensive discussion of those birds on other forums.
And iNaturalist is also much better for identifying unknown birds, either from audio or photos. The system on eBird currently pretty much guarantees that if you post an unidentified bird photo at any level (passerine sp., Catherus sp., Sharp-shinned/Cooper’s Hawk, whatever), you will never get an ID through eBird; you have to post it somewhere else to get it identified. I think this could be improved a lot but it’s just not designed for that kind of interaction. (It’s technically possible: you can report the ID of the bird in the photo and that will be sent to the reviewer, who can email the user who submitted it, but this is sort of discouraged I think)