(There is a takeback provision after 50 years or so, but ...)
Schools and academics can drop subscriptions or boycott submissions. Law can be changed.
Even if that's future-looking only, it can change the balance of interests for journal publishers.
Actually not. This would be likely dismissed before a hearing was held.
In any case, what was asked was whether there is a snowball's chance of hell of nullifying a copyright claim. The answer here is "no".
Good luck trying to change copyright though. I spent about 7 years in DC working on copyright and patent reform. It's very lonely.
> https://archive.org/stream/GuerillaOpenAccessManifesto/Goamjuly2008_djvu.txt
> https://archive.org/details/GuerillaOpenAccessManifesto
If lots of people would do it, there can only few show trials be initialized by the publishers."First they ignore you, then they laugh at you, then they fight you, then you win." - Mahatma Gandhi
So you lobby for an abolishment of the copyright laws - interesting idea...
Indeed: It is quite impossible to claim that "what Elsevier does should be illegal" and not standing up for an abolishment of the existing copyright laws.
On the other hand: The copyleft free software licenses currently only work because of the copyright laws. Keep that in mind...
As you can see, your simple statement has deep consequences. I personally rather think, what many academics do, should be illegal. Here the consequences are far less radical...
EDIT: to agree and amplify your comment:
In this case the copyright is dubious in the first place. Authors who work with public money should not have exclusive copyrights of the work they produce within this arrangement.