This thread includes some comments by IDers about why they might add IDs to observations that are already RG: https://forum.inaturalist.org/t/shouting-into-the-abyss-why-do-some-easy-observations-sometimes-go-unidentified/53506/ (Fair warning: it also includes some rather despairing wailing by yours truly about the difficulty of trying to figure out how to distribute IDs more equally and coordinate IDer efforts in a taxon that is chronically short of IDers.)
My purely subjective impression is that apart from users adding additional IDs as part of their process of learning or systematically reviewing a taxon, there are quite a few users who look at observations via their subscriptions (e-mail or dashboard notifications) or using Explore (e.g. generalist users who like to browse recent observations in their particular region and add some IDs along the way).