Harv, best to use a squish of 1.2-1.3 mm to allow for any carbon build up in the cylinder. If you have the squish too tight, the carbon will start to impact the head, which is not a good idea. In fact, on a large cylinder capacity, we have found more power by increasing not decreasing squish.
Measuring the volumes the way you are doing it is not accurate. It is very difficult to ensure all air is out of the chamber, it does some very strange things as fluid is added! Our preference is to use a plexiglass plate to measure the head volume, where the trapped air can be observed clinging for dear life to the surfaces.
To measure the piston crown and deck volume, move the piston down from TDC, lightly smear a little grease in the bore to seal the ring end gap, then return the piston to TDC again. We then make spacer rings of known volume (calculated from diameter and depth), with the same plexiglass plate greased on top to measure the nett volume. The simple calculation reveals your total intruder/dish volume volumes.
Head gasket wise, measure the compressed thickness and diameter to calculate the HG volume. This is okay as long as the gasket is circular, for those with cutouts, plexiglass with clamps to a surface plate works well.
A very quick calc from your figures above gives; 3900/6 = 650cc per cylinder. 650+63.8 total volume (if that's what you measured?)/ 63.8 gives 11.18:1
To find a theoretical compression volume from a known or desired CR, divide the swept volume by the CR-1. Example, for 12:1 desired CR, take 650/11 (12-1) = 59.09 unswept volume required. Feeding this back in gives 650+59.09= 709.09/59.09 which gives 12:1.
Shout if you need any help Martin.