"I've not got hard stops on the table."
I should have been more explicit. By "hard stops" I meant at the end of the lead screw (threaded rod) which moves the mill table i.e. the point at which the table wont move further. It is the reference point for repeated measurement.
"Doesn’t your method rely on the laser axis being absolutely co-axial with the body? I’ve had a couple of laser collimaters for my telescope, and even with those the dot describes a circle on a distant wall, if the body is rotated in a v-block. "
Apologies for previous brevity again. I don't believe that "being absolutely co-axial with the body" is too critical providing the laser pointer is held in the vice jaws at the same angle to the horizontal (hence shining on the cross lines on wall). The cross lines in the workshop are about 4 metres from the laser source so when lined up, it repeats the original, DTI-based, set up accurately enough. I do check from time to time and haven't found significant change. Below is a picture taken of the laser spot centralised.The image is poor as the laser spot didn't focus well.
One of those many jobs awaiting my getting 'a round tuit' is to mount the laser pointer in a substantial block which would sit within the vice jaws at the correct angle and orientation just to save a few minutes each time.