A key feature of the thermistors you cite is that they are glass encapsulated, and hermetic. Ordinary epoxy dipped thermistors are not hermetic and can fail in DC circuits. This is because the thermistor can adsorb humidity and if the potential across the thermistor is greater than 1.5V, the device will generate hydrogen at one end and oxygen at the other, which fractures the thermistor from inside. The vendors won't be much help here, and don't like to talk about failure of the part. The failures can occur very far into the life of the product, with some shorted, others open, and the balance drifting way out of spec. I learned this the hard way with a medical product and it required a recall of every device and redesign.
As far as the controller is concerned, I assume it is a TEC you are looking to control, and getting an COTS solution is good for one of kind or limited production application. A limitation of a glass thermistor is the thermal lag or time constant because you insulated the part to make it sealed. An alternate sensor is the AD590 family from Analog Devices. It has hermetic metal can contruction, and has a very fast response in comparision.