I think you're going too far here, both in saying something that sounds incorrect, and in reading things into the essay that I don't see there.
There's no need to say that P and ¬P are ever both (somewhat) true, only that evaluating a scientific theory isn't just a matter of finding some single consequence of it to say "this prediction is false: REJECT" (a sort of crude Popperian picture). Instead, the theory should be tested on its overall explanatory power, ability to predict, etc, etc. For each statement the theory implies, that statement is either true or false. But considering two theories, each of which implies some falsehood, we may wish to say that one is still less wrong.
I also didn't notice the sort of Kantian picture you allude to in the essay.
There's no need to say that P and ¬P are ever both (somewhat) true, only that evaluating a scientific theory isn't just a matter of finding some single consequence of it to say "this prediction is false: REJECT" (a sort of crude Popperian picture). Instead, the theory should be tested on its overall explanatory power, ability to predict, etc, etc. For each statement the theory implies, that statement is either true or false. But considering two theories, each of which implies some falsehood, we may wish to say that one is still less wrong.
I also didn't notice the sort of Kantian picture you allude to in the essay.