It's hard to predict the magnitude of the decline, because so many factors (including the release of properties in the shadow inventory, the foreclosure rate, the unemployment rate, the amount of new construction, the oil prices, politics, loan underwriting policy, etc) affect housing prices. Nevertheless, it's pretty clear that the real-estate market correction isn't finished, and prices most likely will continue to decline more in many areas.
Brandon's suggestion works similar to dollar cost averaging for stocks, and it's a method that some investors use to increase the efficiency of their investments. For example, let's say that one sells property A for a loss of $10K, and buys property B for a gain of $20K. The result is a net gain of $10K, and possibly a lower monthly payment (if one acquires B with financing).