This indeed sounds weird - we're talking about the digital output, right?
Sure, there will be some jitter on the Toslink (this is actually actually always the case because of ISI due to the limited bandwidth of Toslink) but a decent and up-to-date DAC will have a fair amount of jitter suppression in the clock recovery.
Did Amir's measurements actually comprise a jitter measurement?
Back in the beginnings of CD, when engineers were not yet aware of the issue "jitter' and the clock was generated at the transport and handed through all stages until it finally was used as word-clock at the DAC, jitter was an issue.
I remembered that I once compared a high-end transport to the SPDIF output of a normal CD player, both paired with a separate DAC. Although being prepared to hear no difference, I did believe I do.
T+A invented the "reverse clocking" scheme (having the clock generation at the DAC) and developers got more and more aware that jitter can be audible. But this was in the 80`s or 90`s...
Question regarding the DAC measurements in general:
The spectrum that Amir shows is measured with a perfectly clean SPDIF or Toslink signal from the AP I assume. Right? This would only show if the DAC itself adds some systematic jitter.
Feeding the DAC with a systematic jitter at different frequencies would show the ability of the DAC to suppress jitter present in the SPDIF or Toslink signal. This would be quite interesting.
Does the AP support such a measurement?
I have hooked the WIIM Pro to my D-6s and I'm perfectly happy. I ordered the Pro instead of the Mini mainly because of the form factor - and the Pro was on discount.