@TiN,
As far as I know, the SR-104 is still considered a primary resistor standard, for two good reasons, first, the jump from a JJ resistance to a 'standard' value resistor is not direct, JJ resistors are very odd values and secondly, they are extremely expensive and compared to a JJR, an SR-104 is very cheap. Because of their reputation, an SR-104 holds its value very well over the years and an older unit often has stability around 0.1 PPM/year and even less. The only time you might find an SR-104 going for less than about $2,000 is if it has been messed up by an idiot, in which case, you avoid it like the plague. For most hobbyists, even the more serious, a very good L&N standard of the 40xx-B series (for example) have very good stability and generally cost less to have calibrated than an SR-104. These can be found for even a few hundred for brand new ones, I purchased a brand new, in the sealed bag, 4025-B for under $100, made in March 1992. It came with a really nice calibration chart for temperatures above and below the nominal 23°C in 0.1°C increments.
The trick is, don't buy a standard that is far above your real needs, the higher grade standard, the more it is going to cost, new or used. I have an SR-104 because I needed it for my work, same thing for the 242D bridge, at times I need PPM accuracy for sure and the SR-104 is calibrated accordingly (that costs more than many other resistor standards). If you seriously need that high of precision resistance readings then you need an SR-104 and an appropriate resistance bridge.....no a 3458A is not a resistance standard in and of itself, although it can be used as a short term transfer standard if calibrated against an SR-104 or equivalent. While it is possible to do a very accurate ratio transfer up and down from a given standard (such as the SR-104), the accuracy and uncertainty does increase a little bit for a 10:1 and you lose a little bit more for a 100:1 ratio, generally that is about the limit for high accuracy ratio transfers, you can do higher but you rapidly lose accuracy and uncertainty increases faster. I use the SR-104 for the reference for 10:1 and 100:1 ratios and I can maintain accuracy and uncertainty to under 3 PPM (carefully), for low ohms, I use low ohm calibrated L&N standards to transfer with to maintain high accuracy.
I think the SL935 is definitely a prototype and a very neatly done prototype as it should be. It is not uncommon for a company like Fluke to build prototypes and try them out before deciding to go further but it is also very unusual for a prototype to find its way into private hands. I would surmise Fluke was possibly attempting to produce a resistance standard that could possibly compete with the SR-104 and provide two resistor values in the bargain. Since an actual product never made it out of engineering, I think they were unable to meet whatever design goals they had in mind. It isn't easy making a standard like the SR-104, otherwise there would be more of them out there at a lower price (perhaps).