Ok, yeah I see it

The first three items (Temp1-3) are nothing, Local and Remote I'm not sure of, HD0-2 are your Hard Drives (you have three? each one of their temps is fine), SYS is your chipset/MoBo (whichever you prefer-they're about the same temp-wise), the PWM's I'm not cure of either, the second Temp1 could be the other chipset (NB and SB), and the Core and Ambient are both on your Nvidia card (both seem to be in the safe temp range).
Honestly, things look good. A bit high by today's standards, but they were designed to deal with the heat, as were the components they are paired with. If you were seeing those temps on a 1.4 Ghz P4, then I'd worry, but for a 3.4 Ghz 90nm CPU, pretty nice.
That being said, I'd still look at your HSF and Thermal Paste, as that is still the most-likely culprit. I believe you when you mentioned it was applied correctly, but depending on the brand/formula applied and given the relatively high heat of the CPU, it could just need a reapplication - it happens a lot, compounds break down and lose effectiveness - kind of like engine oil. This link will show you the specifications for your CPU:
http://processorfinder.intel.com/details.aspx?sSpec=SL7J8. See also:
http://www.intel.com/products/processor/pentium4/specs.htm#. As you can see, the max suggested Operating Temp is 72.8*C = 163*F, the shutoff point for you.
Pull the HSF off and either spray with compressed air between the fins, or remove the fan, brackets, etc. from the unit and rinse the actual heatsink metal itself under water (no soap), wait for it to *completely* dry (air dry is best, put it over a heating vent or something), then reassemble. Take this time to remove all TP from the CPU, then reapply. Just a word of advice, take the 'how to apply TP' website instructions and throw them out the window. Yes, a dab in the center is nice, but most likely, your entire HS makes contact with your entire CPU heatshield. With my Pentium D (another 'oven'), I applied a thin layer over the entire CPU heatshild, THEN placed a small dab in the center. (I also tried both ways the website instructed as well beforehand, to test the entire range of options.) In the end, doing it the way I just described yielded the lowest Temps by a very considerable amount (6*C difference between the next closest) in the end (all three ways were given a break-in period).
Best bet is to get some Arctic Silver 5 (only a few more cents if you buy at Circuit City or Best Buy rather than paying shipping with Newegg). That stuff is awesome!