I agree that fixing a leaky heater valve would be a better solution than an "ugly" by-pass, but this didn't work for me. I had the heater core on the bench when I replaced the valve, and I made sure that it was closing properly. But after all was said and done, I was still getting heat in the cabin. I'm unwilling to tolerate this if I want to drive the car in Florida during the summer, even though my A/C works well. So, I went about trying to find the causes of heat in the cabin.
1. The heater core still warms up even with a well-functioning valve: I've come to the conclusion that this is because the heater core will conduct heat from hot coolant sitting at the closed copper heat valve, (as well as at the heater core output, but to a lesser extent). An effective by-pass will eliminate this, at least at the intake pipe.
2. The firewall: I put Dynamat under a new firewall pad on the engine side. In the cabin, I put a layer of Dynamat covered by a layer of insulation material that I found locally. This material is like miniature bubble wrap sandwiched between foil.
3. Tranny tunnel: Used Dynamat plus the bubble wrap stuff
4. Exhaust pipes: Thermo-tec wrap (thanks Jim), heat shields (stock), and Dynamat plus the bubble wrap material in the cabin on the passenger side.
3. Differential heat: I could feel heat coming from my diff if I removed the parcel shelves after a good drive. Used Dynamat plus the bubble wrap.
4. Less than perfect sealing of closed cowl vent and side cabin vents: At high speeds I was able to improve but not completely eliminate air flow thrrough the cowl vent and side cabin vents using weather stripping. I still need need to close the chrome louvres and swivel flaps to eliminate all air flow at higher speeds.
5. Finding air leaks in firewall (used a light in the engine compartment in a dark garage, thanks Cees) and plugging them with 3M weatherstripping putty.
6. Window "tinting": I selected a material that reduces heat without significantly darkening the windows.
I can say that I have solved my heat in the cabin problem with all these measures.