I think this has been discussed before, but I will try and summarize:
Trying to seat a bullet perfectly, say .005" from the lands is a real feat. And seating exactly to zero jump, without jamming into the lands nearly impossible. Reproducible distance to the lands is very hard to get perfect.
Why? Several reasons.
-- Because you seat the bullet with your seating die with the cartridge base at the max distance, but the act of firing pushes the cartridge forward in the chamber, relocating whatever headspace you created in the sizing process to the rear of the cartridge. Did you account for this in whatever process you used to determine your CBTO? Are your sized cartridges identical to the dummy round you used? And if you did go to all that trouble, remember that your brass will have variable spring back, even if you have annealed. All this amounts to at least a little, say 1-2 thou variance.
-- Then add to that some variation in the bullet itself. The very best are close to perfect, but most are so-so. A thou difference to the ogive is certainly within spec for most manufacturers. A lot more for some.
-- Then add that your seater is almost certainly NOT seating from the ogive, but somewhere further towards the tip. More variance.
-- Then add the reality that over time, your rifle's throat is eroding.
So, if you are aiming for say a nice modest .025" thou jump, a .002" thou variance is just 8%. Not a lot.
And if you aim for a .005" jump, that .002" variance has now ballooned to 40%
Variance is the poison to accuracy.
And if you are trying to get to the lands perfectly, you WILL not get it right. Some will be jammed, some .002" off. The difference is huge in terms of accuracy. The bench rest guys just have a loose neck, jam every bullet into the lands and they get perfect uniformity.
Worthless for a hunting rifle. You must back off in my opinion. I suspect anything closer than .010" is asking for trouble in terms of accuracy. But that's an opinion -- would like to hear seasoned shooters thoughts.