Antman, the short answer is that this is not a problem. There are folks out there, especially overclockers, who insist that a 1:1 ratio between the FSB and memory bus is optimal, but I have never seen any evidence to show why. I believe this is a holdover from older systems.
The way it works is this. The Northbridge divides the CPU bus (FSB) from the memory bus. This allows the memory to run at a different speed than the FSB without any performance degradation. Unless you choose some ridiculously mismatched parts, the memory will always be running slower than the CPU, so there will be times when the processor is waiting for the memory to catch up for certain operations.
For CPU-bound tasks, such as heavy math, the memory speed is not a factor. When doing memory-bound tasks, say sorting or recalculating large spreadsheet tables, the memory speed will be the limiting factor in the operation. For I/O-bound tasks, such as reading data from the network and writing it to the hard drive, the speed of the I/O bus and components are the limit. Each step down this scale is an order of magnitude difference. However, most people doing everyday tasks will have a balance of different kinds of operations.