I disagree.

I think you are contradicting yourself. If a previous work has been copy and pasted, and a novel reader doesn't know, wouldn't the reader benefit from the option to actually read the previous work as a whole?

All credible authors I read mentioned quotes from earlier works. In fact, that is on the one hand an ego boost as a prolific writer, and also helps sell more copies in case of being purchasable.

Most credible university profs in Germany from the 1990th for example always referenced their former work and mention changes of the context, or in case of a theory, modifications.

Books for example, are reprinted and it has been mentioned whether changes to the content has been done.

Personally I really see no problem, leaving the decision, whether you copied something or not, to the reader.

Many forums have/had policies about not doing cross posting (in different categories). I find this similar.

Yes, maybe from the "plagiarism" angle is not very relevant, but I would prefer not to have a system in which people try to "flood" repositories (journals, etc) with the same thing over and over. People looking for new information, people reviewing will get most of the burden to "keep things clean" while for the poster that is not a problem.

If they don't know about the previous work, they won't know to go there for more. If they do, it doesn't mean they instantly recognize it and know to skip.

Just mark it, it'll take seconds.