I agree, but that doesn't rule out non-local interactions in general
What these party-pooper theorems do is largely rule out being able to differentiate non-local interactions from local interactions... which doesn't mean they don't exist, but if they do, there's large pieces of the puzzle missing before we can show they do.
If there is a non-rude way of conveying the message “when I said that it depends on the definition of interaction you gave me one and then you told me to ignore it” please assume that I had written that instead.