So as I understand you want to add effects on the "live monitoring" while not adding effects on the recording.
This can be done.
You just need to run the "live monitor out" of your yeti microphone through a reverb effect, and then to your headphones.
So how can you do that easily?
The simplest way is to just buy a hardware reverb, and run the live monitor of your yeti through the reverb.
Something like this would work, but you can most certainly find cheaper (or much more expensive :P)
http://www.sweetwater.com/store/detail/MX200
And alternate approach would be to run software reverb. This requires an audio interface with multiple in an out. Because you are using a USB microphone, I assume you do not have an audio interface, so this solution is probably not idea.
Anyways, you could run the live monitor of your yeti into an audio interface, into a track of a DAW software. Then you can place any effect your want on it, then live monitor that specific track in your DAW.
The problem with this, is that you will need to keep your ASIO buffer very small (64 samples) so you don't introduce latency into the live monitor. If you are running two many effects (or too complex), your cpu will lag and you will get buffer underrun (pops and clicks).
My opinion is that you get a cheap hardware reverb and run the live monitor of you yeti though that and to your headphones.