A passively cooled switch only using half the wattage would double the battery life too which would be great for a slim portable version.What if Nintendo does have 2 models launching in the next year and a half. What if they actually wait til Spring 2020 to launch the Switch Pro, but the holiday before that, they launch a Switch Slim pushing the current Switch's performance on the go, could also have the new SoC from the Pro model for $200. One thing they can do is make the dock cheaper, the Switch itself could output the signal, and the dock would just be a pass through with USB ports. They can make the dock much smaller too. They can make the tablet ~9 to 10mm instead of 14mm thick and make it passively cooled. The rails would remain the same size and the front of the device would still sit flat with the joycons, but the back of the Switch Slim, could have gentle curves towards the rails, so when the joycons are attached, the back of the device has a slightly wii u gamepad grip. The Pro could then come out in a quiet quarter to push sales, while the slim could help drive the lower price towards kids.
As for technical ability, 16nm consumes 60% less power than 20nm, so the 9 watt SoC when docked, should drop to just 3.6 watts, plus another 2 watts from the rest of the Switch. 12nm is a free TSMC upgrade and has less power leakage, further reduces power consumption. This should bring such a device down below 5 watts and that should make it possible to cool passively, especially given the size of the device.
But I think Nintendo should just keep two SKU's. An entry level one (launch Switch plus the 16nm X1's) for $199 and a premium model with the new Mariko chip for $299.