Better the enemy you know than the one you do not. That false?

And how is it we know as much about closed models as we do about the open ones?