This is a business not a technical question. How much money is your company prepared to invest in legacy software? Is there quantifiable benefit?
I tried porting a moderate-sized C# asp.net 1.1 website to .Net 2.0, and in the end I couldn't justify continuing; I had too much rework and too much re-testing. It wasn't worth it from a business and risk POV. The website was turning over > 1M USD a month -- making my life a little easier just wasn't worth it; I couldn't put together a business case that made any sense.
You know your situation better than us, but I would suggest you consider the timeless adage, if it ain't broke, don't fix it.