I have yet to see a good explanation of what advantages a .18 micron process gives. Canon has the capability, so apparently they have not been seeing a compelling reason either.
My first thought was that wiring traces that must run between pixels could be made smaller, and on chip transistors could be made smaller, but I have read that it is not possible to make the ground and power traces smaller due to crosstalk and voltage issues unless you adapt Canons new patent for a rear illuminated sensor. The transistors could likely be smaller, so thats a incremental improvement. Using more of the available surface area for photocells is obviously a good thing.
I'm certainly all for new technology if it improves the sensors, but not for new technology just for its own sake.