Mr. Stern is a very intelligent man. Several years ago, when I was in the Mopar newsgroup for Chrysler items, he was in there too. He can be a little opinionated sometimes, but knows what he's talking about. He also has a way to make a late model Mopar that has a computer controlled alternator system into the old way of using an external regulator (like for his '68 Dart from NAPA) and how to fool the body module computer into thinking everything's all right--quite interesting.
I know that it's always better to use a relay to control sizable amperage circuits, BUT why are the headlight switches cooking if there's a circuit breaker in the circuit too? In the case of the '72 era Chrysler C-body cars, the circuit breaker was external to the switch. Other cars have the circuit breaker in the switch. As it isn't a fused circuit (in a safety-related item like headlights), there will surely be a circuit breaker in the system somewhere.
Seems like I recall a discussion about using a relay to run the headlights so they'd get full system voltage and resultantly be brighter. I also recall some people trying to put higher wattage bulbs in their plastic headlight housings and the housings melting from the heat.
Could it be that the headlight switch (with the supposd 30amp breaker) could be situated in a place on the instrument panel that gets little air circulation or is a poor heat sink? Or there was an inferior grade of lubricant in the switch from the factory that let internal corrosion happen? All the headlight switch is doing is transferring current with little or no resistance, but touch the connector on the back of the headlight with them working and that's where you'll find some heat. If, perhaps, the contacts are too small for the current being carried, come resistance and heat can result over a period of time, especially if the lube goes away and internal corrosion happens, but still if the circuit breaker is working, it should kick out before significant heat happens, I would think.
Considering the financial situation at Chrysler in the latter-middle 1970s (when the 1980s cars were being finalized), it seems that many of their product-related problems were due to decisions related to cost instead of just poor designs (as many desired to believe). LCD displays take very little power to run, just as LEDs take basically zilch power to run. The less heat produced from an electrical device, typically the less power it takes to run it.