> arxiv.org has had the best exposure it's likely to get in years. The more that publish there, the fewer who publish behind firewalls
We're talking about condensed matter physics here, arxiv needs zero exposure in this field. It is extremely common to publish on arxiv first and only then begin the process of submitting the same manuscript to be published (and peer reviewed) in a journal.
And while there are still some authors that skip over publishing pre-prints at all, there's no serious arxiv competitor if you do decide to publish pre-prints. It is a de-facto monopoly in this field.