The alternative is that you think people should be able to use social media platforms in ways that violate their rules, and that the platforms should not be able to refuse service to these users. I don’t think that’s a justifiable position to take, but I’m open to hearing an argument for it. Simply calling it “hellish” isn’t an argument.
And can you clarify if your position accounts for spammers? Because as far as I can see, your position is very clearly “spammers should be allowed to spam”.
Of course in the ideal world all bans would be handed out correctly, be of a justified duration, and offer due process to those banned. We don't live in that world, the incentive is emphatically NOT to handle appeals fairly and understandably. Getting truly permanently banned on a major platform can be a life changing experience.
In reality users can generally get away with signing up new accounts, but new users will be marked somehow and/or limited (e.g. green names on HN) and get extra scrutiny, and sign-ups will have friction and limits to let it not scale up to mass spammer scale. The rest is handled manually by moderation staff.
The limits to moderator power are a feature that compensates for the limits to moderator competence.
This is an absolutely gargantuan-sized antifeature that would single-handedly drive me out of the parts of the internet that choose to embrace this hellish tech.