I might not have much time to continue here in the next few days so I'll start by thanking you for the interesting and thoughtful discussion.
I do agree with some of your important points. It is a problem that copyright works by creating an artificial monopoly without sufficient control on how that monopoly is then used. It is a problem if there's insufficient incentive for creators to make works permanently available to those who want them. As I said earlier I do think some reform is needed in this area.
I don't think that's sufficient today to support your central argument that works should only be available on a permanent basis. The two financial models we've been discussing are a trade off between having access to specific works indefinitely (the sale model) and having access to many more works but only for a limited period of time (the library model). Either way the scope is limited and that makes the price viable. I don't think the prices for streaming services that operated on the alternative basis you're proposing would stay the same. I think they would go way up and that could reduce a lot of people's ability to access the long tail of works that they wouldn't want to spend the full price to buy permanently.
I also want to point out that your views here appear to be based on the US implementation of copyright. Some aspects of copyright in the US are at least questionable under the related international treaties already but pragmatically the US is powerful enough economically that its big companies can get away with a lot for now. Likewise the US itself can get away with having very generic fair use exceptions in its laws. That could change - with very significant economic effects - if the US attempted the kind of legal shift you're proposing without engaging strong international support first. And that would obviously require convincing everyone that the economics would continue to work. In any case some of your arguments about backups and physical media wouldn't apply in most of the world (or even in the US given the DMCA - but of course DRM and the legal protection it's been given around the world is a whole controversy in itself because it potentially makes a lot of the legitimate criticisms of copyright even worse).
IMHO your last point is the place we should look first. The problem is that the market isn't really competitive when copyright holders can license the works exclusively. This leads to problems like works only being available on a single streaming service or broadcast on a single TV channel. Sometimes different seasons of the same show or successive films from the same series end up on different services or channels. Sometimes works get yanked from a streaming service's library with relatively little notice when their deal runs out. If we changed the rules so that works being offered commercially to one distributor also had to be offered to other distributors with similar terms and pricing and within a relatively short time then I think that alone would restore a lot of healthy competition in the market and solve some of the significant problems with how the current distribution models are working. And importantly it is relatively practical to explore options in this area compared to trying to radically restructure the entire copyright model globally.
There's a lot more that we could discuss here. For example we haven't even touched on creative works like software that tend to be updated over time and the different models for supporting that. We haven't considered how to deal with work of potentially high value but a short useful lifetime like news reporting and investigative journalism either. It's a huge area - and that's a big part of why I think throwing out the whole principle of limited licences in favour of allowing only permanent sales is probably not the right way to go here. But the problems are real and I do agree with you that the system could be significantly better than it is today.