It is a tricky subject. When I referred to a model I had published elsewhere, I was asked to write about the model again in the current paper, since readers may not be able to access or want to access the other paper. But more recently, I was asked to do the opposite, since the model I was writing about in the Methods section was the same I had published elsewhere.

I see no reason to change terms and structure in scientific publications so as not to plagiarize myself.