70 uL of blood is usually more than enough for FACS analysis of transplanted NSG mice. For setting the panel (single stained, FMOs, etc..) you can take about 5-7 uL of blood from each mouse sample (if you have 30 = 150-210 uL total), pool them together, and resuspend them in a volume suitable for FMO staining (10 colors, 50 uL each >> 500 uL). You can avoid irrelevant FMOs, if the staining is clear enough (ie. black or white antiigen expression, such as hCD45 vs mCD45) and dedicate more cells for markers with more continuous expression or markers staining only a small subpopulation. For single positive controls, you should use stained beads (ie. BD Compe Beads). I would not use human PBMC as the cell subpopulations and autofluorescence will most likely be quite different from the samples you will be acquiring, even if the mice are engrafted with human cells.