The concept you are having a problem with is simple...Series circuit resistors INCREASE RT ..Parallel resistor circuits DECREASE RT. .And b/c the sensor decreases as it heats, by lowering its R with another parallel R, the R factor triggers the fan EARLIER than normal temp spec. If the CTS was not a NEG type sensor, the R bridging would not work. You are lowering the RT of the sensor before its normal heat/R relationship...in other words , you are taking it out of spec to the low side. The ACC panel triggers the fan when it sees a DECREASE in R, not an Increase. And b/c you have lowered the RT, it reaches the cut-in R spec earlier [ @ a lower temp], thereby fooling the ACC into thinking the temp is 105C, when it is really NOW a lower temp than that [ depending on R factor of bridge R used]. You have not changed the ACC panels trigger spec..you have changed the sensors temp/resistance relationship value...ie, taken it OUT of Factory spec. ya broke it to your liking ......
There is a Factory spec. chart for Temp/Resistance values for the CTS ...with the bridging R you have, your sensor would actually Fail the factory value test...but that is why we do it...and with a little math, we can change the sensors value to come on at about any temp we want.
Remember, the bridging R is NOT another thermistor..it is a set value Resistor.
http://physics.bu.edu/py106/notes/Circuits.html