You can do it using
SDL_MixAudioFormat() but bear in mind that the two WAVs that you mix together need to be in the same format (same sample rate, same number of channels, same number of bits per channel). Whilst you could do a format conversion in real time using
SDL_ConvertAudio() personally I would always prefer to do the conversion ‘offline’ and supply matching WAVs with the app.
What becomes more interesting is mixing a second WAV ‘asynchronously’ with a currently-playing WAV (such as adding sound effects on top of music) without using SDL_mixer. For a long time I thought that was impossible, but it can be done with the proviso that there’s a race hazard (I asked a question about this here recently). It also seems not to work properly when using WASAPI in Windows so I force the audio driver to directsound in that case.
It’s amazing what you can do without using SDL_mixer if you try hard enough!