This is so terribly wrong!
You're making the typical mistake of equating core count to CPU performance. That is a common mistake, because that is exactly what you would expect to see in the PC space. As such, it isn't surprising that all the examples you site are from the PC world, where CPU's are explicitly designed to work that way. This is nothing but a consequence of Intel's and AMD's business models. Since PC CPU cores within a generation are identical, fewer cores = fewer transistors = lower manufacturing costs = CPU's that can be sold at lower prices but potentially offer less performance (although depending on the software used there may be no performance difference at all).
However, there is no law in physics or integrated circuit design that states it must be so!
CPU's in the mobile space are entirely different, as are the business models. You start out with a transistor and power budget and then go from there. If you have the same numbers of transistors to invest in either a dual-core or a quad-core design, the question of which will perform better is a very tricky one. Here, all your experiences from the PC world break down completely. The performance difference between the MSM8960 and Tegra 3 is just one example. The later is a quad-core, whereas the former is just a dual-core, but it is the much better performing CPU regardless.
I don't agree with the notion that we don't need more powerful hardware. As long as battery life isn't reduced any further (it's bad enough as it is), we should welcome any and all further performance gains.
However, we do need to abandon this false assumption that more cores = better performance. It's far to simplistic and in the mobile space, although it can be true, it doesn't have to be.