Elsevier muddying waters, the 36th
This seems to be a new not-quite-policy (the specific page was first crawled by archive.org in August 2018, so I think it is new):
In all cases accepted manuscripts should:
[...]
not be added to or enhanced in any way to appear more like, or to substitute for, the published journal article
What does this even mean? Of course accepted manuscripts are a substitute for published articles; no one is downloading Elsevier PDFs to see their slick formatting or cool typefaces!
This seems like another case of making rules so confusing that people who are inclined to risk aversion won't bother to use their rights.