To me, the issue is not whether the "laws" emerged slowly from other laws, if we are to presume that would require having other laws about how laws emerge. It is how well can we rely on the language we use about laws "governing" behavior in the first place. I think this is a shortcut, that often works fine, but what evidence do we have that this is the reality? When we find a "law" that works, we tend to say it "governs" reality, just look at the language of "laws" and "governors"-- it's classic personified imagery (we interpret according to our familiarities). But then when its limitations are discovered, we conclude "it was only approximately the right law, there must be a deeper law that is the real one". As if we found a Lieutenant but still seek the General. Why would we not simply conclude that we have noticed patterns in past behavior that is predictive of future behavior? The elegance is in organizing the patterns in the most mathematically clear way, not in finding why reality does what it does. At what point are we not organizing patterns in our familiarities, but instead figuring out what makes reality work the way it does? I might go even farther than Wheeler: never. It's just not what we're doing, it's only what we're pretending to be doing.