Yes, pretty good article. The only problem with it I caught is that the implication that diacetyl is rarely formed outside the cell (i.e. by non enzymatic oxidation) is counter to the usual thinking which is that diacetyl is formed after the yeast have been removed unless all the acetolactate has been removed i.e. it is, for the most part, formed by non enzymatic oxidation. This is the basis for George DePiro's simple test: warm a sample of beer you are about to package (at which point, presumably, you have removed the yeast). If acetolatate is present the warming will speed it's conversion to diacetyl. Thus if you smell diacetyl after warming, you can pretty much count on it forming after the beer is packaged and should krausen.
This model is also responsible for the efficacy of Maturex (alpha acetolactate decarboxylase) which catalyses the conversion of acetolactate directly to 2,3 butane dione.
If you want the last word on diacetyl get a copy on Inoue's monograph "Diacetyl in Fermented Foods and Beverages". More than the average homebrewer probably needs to know but some pretty interesting information in there (Japanese people hate the taste of diacetyl).