The big point I took away from this article is that wanting and liking are different, and thus we should be skeptical of "revealed preferences".

But the title seemed to imply that the article wanted to address the question of whether or not we should wirehead. The last paragraph seems to argue that we should be really careful with wireheading, because we could get it wrong and not really know that we got it wrong.

Go too far toward the liking direction, and you risk something different from wireheading only in that the probe is stuck in a different part of the brain. Go too far in the wanting direction, and you risk people getting lots of shiny stuff they thought they wanted but don't actually enjoy. So which form of good should altruists, governments, FAIs, and other agencies in the helping people business respect?

I agree with this, but given that it's a central argument of the article, I think it could use a longer explanation.