In scientific work there is more to plagiarism than just crediting people for the work that they do. But first, an aside in the difference between copyright infringement and plagiarism.
Suppose that I find a publisher (unlikely) who will let me take an early edition of Goethe's Faust and publish it under the name Buffy St. Magnus, with no indication anywhere that it is from Faust. I can do that legally, since the work is not under copyright, though some recent editions may be. I don't use quote marks or anything, but even if I do, it is plagiarism. Plagiarism, but not copyright infringement. I'll be condemned (well St. Magnus, anyway). But I can't be sued.
Suppose, alternatively, that in writing a new book, that I take a chapter (not the entirety of the work, but a substantial part) from a recent novel of, say, Louise Erdrich and incorporate it into my own work. Suppose that I'm clear, using quotes or whatever, and giving proper citations of the original. Since I cited it (quotes or not), it isn't plagiarism. But whoever owns copyright to the original will likely object and I'll be sued, almost certainly, for copyright infringement. So, copyright infringement, but not plagiarism. This would also be the likely outcome if I make small changes or paraphrase parts of it.
In the latter case, if I do the same thing but don't give a citation. Then is is both plagiarism and copyright infringement and I'll be both sued and condemned.
But, suppose that what we have is scientific writing instead. Note that scientific papers have a context, usually expressed in their own citations and their bibliography. There is more here than just who wrote it, there is the complete scientific context in which it was written. A scholar reading the work, sometimes need to explore that context directly. If the paper makes that impossible, by plagiarizing another, for example, the chain is broken. St. Magnus says the buck stops here. But really it doesn't. If I just copied from Shi Sen, then a reader of my work has no easy way to find the original and thus explore the context expressed there.
That is what makes scientific plagiarism special.
If it were only about attribution, then self plagiarism wouldn't be a thing. I'm the source of the words, even though I used them before. But if I don't cite myself properly on the prior use, others are cut off from the complete context.
Also, if were just about attribution then putting it in quotes and providing an incomplete citation, say, just the author's name, then it wouldn't be plagiarism either. But most would consider such an incomplete citation to be wrong for some reason, whether related to plagiarism or not. Ancient authors are often quoted this way, of course, but not living (or recent) ones in scientific literature.
To avoid plagiarism, or self plagiarism, you need to cite and you need to be clear about what you are "capturing" from the early work, just to help other scholars find that original context when needed. You can do this in many ways, but you need to be clear about it. Putting a few words of others in the middle of a sentence that I otherwise write is a situation that makes quote marks especially useful. For a paragraph, I might want to indent instead, with the citation immediately following (or at the beginning). But it has to be clear.
However, if I cite properly, avoiding plagiarism, but still copy "too much" (a value judgement) then I'm still open to a charge of copyright infringement. Any copyright holder can make the charge and file a suit. It is then up to others (courts, juries, ...) to decide whether I overstepped the bounds.
But note the important difference. It is lack of citation that makes it plagiarism. It is giving the impression that the words, ideas, whatever, are mine, and not those of the originator. It is the appropriation that makes it wrong, not the lack of "quoting".
That said, quoting formally, can help you make it clear what is yours and what is not and that clarity is important.