They're based on objective visual observation.
Mine are based on a full size model based on an accurate scale map of Dealey Plaza blown up to actual size and actual survey data. I created a full-size model of the car using the H&E drawing and measurements and photographs. I can then make all measurements on the model. How do you determine angles and distances?
It appears some of your "actual dimensions" made no allowance for perspective.
The software takes care of perspective. The perspective in Zapruder's film is somewhat flattened due to the use of a zoom lens.
Observation and perceiving the effect of perspective on measurements and sight alignments.
Actual measurements are better.
 Possibly Connally is centered on his jump-seat in this picture | |  Was the jump-seat itself a good distance from the car interior? | |  Was the front of the jump seat near the rear edge of the grab-bar (looking straight down)? |
That's why I could no longer add figures and LoS using the 2012 diagram. There may be more adjustments and nudging in the future if new photos and the actual 100-X plans surface. I'm going where the evidence takes me.
What are your values for the horizontal and vertical angles at z197? No matter how you do it, the bullet goes right to left
Given all the things you've missed, you started off your SketchUp model with a limousine drawing that was inaccurate. And since you don't know how to do 3-D, you're building up from it was a disaster.
Sketchup handles all the 3D stuff so one does not have to "know how to do 3D". One just has to know how to use Sketchup. As you have pointed out, the scale drawings of the limo by Hess & Eisenhardt are not completely accurate. That is why my early models of the limo are not quite accurate. Using a variety of photos of the limo at different angles is the only way to get the placements of the seats correct. There may be some details that are still incorrect but I don't think they are major ones. If you can point out specific errors I will check and correct them if you are right.
I know you think your 3D model is the stuff of legend, but is there anything in particular that I pointed out about your 3D model that was inappropriate?
Your trajectory alignments at Z197ish and Z270ish mainly conform to your pet theory.
You are using early (2012?) versions of the limo model. Use the ones I have posted. Trajectory alignments are based on the actual straight line from the SN to JFK's back where he appears to be in the car in z193. The position of JBC's left leg is a bit of a guess. But it seems to me more likely that it was out to the left a bit than turned inward to the right. The path from JFK's throat to JBC's left thigh just works naturally.