While the DAC has a tube-based output stage and therefore might have a relatively high output impedance and consequently drive capability that is somewhat limited, the 100K input impedances you mentioned for one of the integrated amps and one of the headphone amps represent loads that approach being negligible. So I would expect that if the DAC can drive each of the other two units individually with good results, and assuming that none of the interconnect cables would be unusually long and therefore the two sets of cables won’t in combination present the DAC with unusually high capacitance, you are very likely to be ok splitting the signal between the LM34IA and either headphone amp, and between the Tube Technology amp and the Feliks headphone amp. However without knowing the input impedances of the Tube Technology amp and the KGSSHV, as well as the output impedance of the DAC, we can’t predict what would happen with that combination in terms of sonics.
Also, a caveat that occurs to me is that some of the input impedances could conceivably be different when a component is turned off than when it is turned on. So if you proceed with splitters it would be a good idea to verify that when you listen via speakers the sound is the same regardless of whether the headphone amp is on or off, and when you listen via headphones the sound is the same regardless of whether the integrated amp is on or off (with the integrated amp’s volume control turned all the way down, of course).
Regards,
-- Al