@digitalfiz, I agree that different certificate authorities perform, and claim to perform, different kinds and amounts of verification.
But as other people have described, many uses of certificates experience a weakest-link phenomenon. HTTPS client software accepts any kind of cert from any CA. So the question @joel is pointing to is to what extent users will check what kind of verification a particular entity was subject to. If they don’t check, directly or indirectly, it’s not clear that they got whatever benefits were to be had from the greater verification, especially since if an impostor could get another cert from a different CA, those users might accept it just as readily as they accept the genuine one.
To put @joel’s point another way, the extra verification is meant to benefit the relying party, the person who accepts the certificate. If it’s extremely rare for relying parties to check this information, or act on it, or become aware of it, how much benefit did it actually get them? It didn’t improve the cryptographic protection of their information, and in many scenarios it wasn’t even likely to prevent an attack based on some other CA misissuing a certificate.
New technologies to address misissuance, like CAA, CT, and HPKP, are applicable to DV certificates too: a site that uses a cheap CA that performs less verification can take the same precautions against other CAs doing the wrong thing as a site that uses an expensive CA that performs more verification.