One of the main issues with mixes based on samples and convolution reverb is a lack of definition and/or muddiness. I will not say it’s necessary for every mix, but chances are you can improve yours by reducing two frequency ranges: Place a -3 dB marker at 250 Hz and one at 500 Hz (Q=1) and then compare the results. Does it help? Let me know!
My approach: I always reduce the 250 Hz for better clarity and most of the time also the 500 Hz range for some “punch”.