I do not have any fixed guidelines except to say that this is essentially a volume calculation. If you know the size of the pin and the hole; then, it is a simple calculation for the volume of metal to fill the hole.
Furthermore, you need to allow for a fillet (I use 20% as a rule of thumb) to get the volume of metal you require. You will then need to double this because solder past is only approximately 50% by volume. This will give you the volume of paste you need to apply. A 1.6mm thick board will probably require a step stencil to get sufficient material deposited to fill the barrel.
A further point is that if you simply fill the barrel with paste much of this will get pushed out when you insert the component; usually this will drip off the pin prior to reflow. It is often advantageous to put a bar across the center of the hole (typically about 30% of the diameter), this forces the paste down the edges of the barrel and reduces the tendency for push through.
The Department of Systems Science & Industrial Engineering of the State University of New York, Binghamton and the Surface Mount Technology Laboratory of Universal Instruments Corporation did an extensive study on the pin-in-paste process using a DEK screen printer and Vitronics Reflow oven.
For more information on "Solder Paste Tests for the "AART" Process" (http://bit.ly/8YzhU2) you can contact George Westby, Director SMT LAB for Universal Instruments, at westby@uic.com. He is one of the authors of the report. Jeff Schake from DEK is also a great contact for this process, he can be reached at jschake@dek.com