@Ryan_Carlyle Stratasys, there’s a lot of marketing wank in those systems? Those are multi-thousand dollar machines, and off topic from the original post talking about cheap printheads.
@MidnightVisions bro, our FFF printers are built using expired Stratasys patents. The original RepRaps were printed by Stratasys printers, largely copying Stratasys designs. Do some research.
@Ryan_Carlyle Sorry bro, but none of that was mentioned in the post, and I’m not a mind reader. Regardless, the topic has strayed from the post.
@Ulrich_Baer I had not seen the Merlin hotend before. Looks like the thermistor is placed closer the the nozzle (good). With the short melt zone, looks to be optimized for small/detailed prints with very small nozzles … at which point I completely lose interest. 
My interest is in printing larger objects, fast, and with sufficient accuracy. A printer optimized for very small/detailed parts is rather a different beast. Enough so as to call for different design.
@MidnightVisions When ordering nozzles, there just is not enough information to make reasoned choices. so I punted and ordered this set. (Have to start somewhere…)
Pretty much the reason for the above post.
Was focused elsewhere after posting, so responding late.
First, I am mainly in agreement with what @Ryan_Carlyle posted above, with the exception that I think measure of filament temperature should be as close to the point of extrusion as practical. If everything else is controlled, the temperature in the melt chamber should be calculable from the downstream measure.
The inverse argument could be made, that you could calculate the temperature at the tip from the measure upstream. But not all the factors are well controlled. We can tolerate upside variation in the melt zone (short of chemical breakdown) but less so variation at the point of extrusion. The end-to-end goal here is plastic extruded at an exact temperature.
With present software and hardware, at present slower rates of extrusion, a large thermal mass in the hotend is generally a win.
In future, at higher rates of extrusion, you want heat delivered at the instant the rate of extrusion changes. (How does nichrome wire fare rubbing against molten plastic? What other resistance heaters have been tested?)
Yes, this does wander a bit from discussion of buying nozzles, but is strongly related. As we push performance, we need measure near the end point. We need to understand the upstream character to choose design of the nozzle, at least a bit.