A couple things: 1) They do indeed vet reviewers, both their reviewers and also who ends up on their editorial boards, which for most people on them is just "Y

A couple things: 1) They do indeed vet reviewers, both their reviewers and also who ends up on their editorial boards, which for most people on them is just "Yeah, you're going to get more review requests." Editors should - but often don't - also evaluate the quality of the reviews themselves. 2) Some journals do copyediting and layout. Several times this has markedly improved the paper, including one that remade a figure. 3)They also handle some backend stuff. For example, for most medical journals, you just have to check a box saying "This was funded by the NIH" and they'll handle putting it in the appropriate repositories, etc.