It's an interesting set of reasons, but I prefer Keccak/SHA-3 over SHA-256, SHA-512, and BLAKE. I trust the standards body and public competition and auditing that took place - more so than a single author trumpeting the virtues of BLAKE.
Ironic, because the final NIST report explaining their choice mentions that BLAKE has more open examination of cryptanalysis than Keccak as a point in favor of BLAKE.
