It depends what functions you are using. for example 3D ray tracing in Photoshop* uses every available logical core but some 2D filters do not. As Dag said - scratch disk and RAM will have the biggest impact rather than extra cores.
*If you are going to be doing a lot of 3D ray tracing , I would advise using a 3D application that uses the GPU for that function where you can utilise thousands of cores and cut rendering times accordingly.
In most normal situations the question is irrelevant. You're not waiting for the CPU; you're waiting for the data to be read from and written to the scratch disk. The tests above are flawed; they were conducted with a single SATA 6GB/s drive, which is slow enough to mask the real result. In addition only 8GB RAM, so not enough to help.
If you have lags and performance issues in Photoshop, that's not the reason.
Yes, some filters are CPU-intensive, but more and more of that is moved to the GPU these days.
(edit: ...oh, and why is it called SATA 6GB/s, when the real throughput is at best 450 megabytes/sec?)
OK, but that's still only 64GB, not nearly enough to contain a real-life scratch file. And in any case it's always written to scratch even if it could be contained in RAM.
So the point still stands. Photoshop is not CPU-limited in most normal scenarios. It is multi-threaded, it does use 8 or 16 cores in parallell where possible (think nine pregnant women) - but that's just not what you're waiting for.