The transform format is a string that is a sequence of transforms, so you can have several after each other.

Edit: Snap doesn't make a distinction between upper/lower case these days, so this part does not make a difference (it maybe worth being aware of it though, in case you see some Raphael.js code and want to understand), the rest should still be relevant though...

T/t = Translate (t is relative, T is absolute)
R/r = rotate(r is relative, R is absolute)
S/s = scale(s is relative, S is absolute)

Its worth looking at the Raphael transform documentation if the Snap.svg doesn't have enough information, as there is a lot of overlap.

For transformations, some will reference a 'centre of origin' about which to rotate/scale etc, as sometimes you may want the centre of origin to be the object itself, sometimes 0,0, sometimes around a specific point.

t-10 0 s0 32 32 would translate x,y -10,0 and then scale x,y,cx,cy so scale 0 on the x, 32 on the way around cx 32.

r180 32 32 would rotate 180 degrees around point 32,32. You can normally use a comma or space to separate values.

after represents attribute values to set after the animation finishes. animafter represents animation values to set after the animation finishes.