As far as I can see, the only things they now provide that hasn't been replicated as a 'free service' is matching editors and referees to submissions, and journals of repute, i.e. publish in Nature and it's a bona fide article, publish on your blog and no-one knows who you are or whether your work was refereed.
To be clear Sci-Hub merely provides copies of work that has been accepted for publication in a journal, it's not a free replacement for the full 'service' that Elsevier and other publishers provide.