Maybe it helps to go into the details of how TotalMix works.
I fear you are used the old-fashioned way, to pull a fader on a mixer, and then you hear that signal in your monitoring.
Perhaps not exactly this, but similar to this.
With TotalMix FX, an audio signal is not directly sent to the output but to the middle row; SW playbacks = audio originating from the PC. Be it Windows Sound, from Youtube, from the DAW.
This all appears at first in the middle row.
TotalMix FX works output/submix oriented. And it works now like a patch bay.
In submix mode, you select an HW output of your choice from the bottom row.
This output has an "individual submix".
You see what is being routed into this submix according to the fader positions from the top and middle row.
Top Row = HW inputs = audio from sources that are directly attached to the recording interface
Middle Row = SW Playback = audio originating from your computer.
Latency-wise, audio from the top row can be routed "on the interface" from input to output port in Near-Realtime.
A typical use case is a monitoring flow for a vocalist singing in.
He hears the backing track in the headphones, and sings, and he needs to hear himself with the least latency.
This you can achieve by routing the Mic input directly to his headphone without a DAW in the signal path.
Audio, which is being routed from SW playbacks, always has more latency, as audio has to go over USB (or alike).
And even worse, if you increase buffers, the latency can become big.
So, this submix / output-oriented routing approach makes the most sense, as you can control from where you get audio. Either in Near-Realtime from an HW input or with more latency from the PC via USB.
Therefore, you need to become acquainted with submix mode to think about routing from the output.
Select an output and now choose what signal needs to be sent to this ouput.
Either from the inputs (top row) or from the computer (middle row).
And if you think about this further, it's fantastic that you can configure routing individually for each of the outputs.
You only have to change the way of thinking. Not simply "I want to hear this and that signal and move the fader up".
First, you have to ask yourself for what output. At first you have to select the output.
And once you become acquainted with it, it is so easy and wonderful to work with TotalMix.
Just select the output in the bottom row, then inspect the routing according to fader positions
and then move a fader up or down, depending on what you need to hear.
You can also think about whether you can catch this signal directly from the inputs (in near-real-time).
Or to get it from DAW (because you want the signal processed).
There is an additional trick for you. You see the little button "sub" (in TM FX 2.x "submix").
You can click / toggle it.
Then you see only those channels from the top/middle row that actually route sound into the just selected HW output.
I hope this helps you. Enjoy!
BR Ramses - HDSPe MADI FX, M-1620 Pro D, 12Mic, UFX III, ADI-2 Pro FS R BE, Nuendo 15, Win10 IoT Ent