I understand this sentiment but don't know what to do with it. "Not even Schneier can be trusted" is "not even wrong". Schneier has very little to do with modern cryptography! But a long time ago, someone created a "Bruce Schneier facts" meme site, and now it's like an article of faith that he's a cryptography engineering expert. No, not so much, and I don't think he'd disagree.

He's a perfectly nice guy with a lot to say about information security and its intersection with public policy. But I think it's been plural decades since he basically declared himself outside of modern cryptography (you could call it at the point where he said he didn't "trust the math" of elliptic curves, which he left out of Practical Cryptography, over 26 years ago).

It's not so much that you should or shouldn't take "Applied Cryptography is bad" personally; rather: if you think Applied Cryptography is a useful reference or learning tool, it's pretty important to know that it is not.