[Cicero], and Harry Caplan (1896-1980). Ad C. Herennium de Ratione Dicendi (Rhetorica Ad Herennium). Loeb Classical Library, 403. Harvard University Press, 1964.

It shouldn't escape the mnemonists' attention that while Wozniak recognizes some basic attributes of memory and mnemonics, he obviously isn't steeped in the traditions of the art of memory. Specifically he doesn't seem to be aware of associative methods beyond peg systems and some low level basics which he's come across via Tony Buzan. He's also missing the major system and the method of loci in general. However, when looking at his list of "Twenty rules of formulating knowledge", the majority of the items on the list are either heavily informed by the memory canon within classical rhetoric. Some of the items on the list actually move in an opposite direction from good memory principles (his admonitions against sets and enumeration), but it's because Wozniak is explicitly missing some of the basic mnemotechnical tools.

Have any writers on space repetition gone beyond Wozniak's state of the art with respect to mnemotechniques before?

I had started it and lost it due to a technical glitch, but it might be worth highlighting the places where Wozniak's list either directly dovetails or diverges from the arts of memory. His list could be dramatically improved and compressed by brining it closer in line with the fourth canon of rhetoric.