I didn't want you to think that your question has been ignored ... it's just not a simple one to answer, as you've noticed from the answers you have received to the same question posted on other cooking forums. And the more I tried to find a simple answer, the harder it got.
Dextrinization is apparently not necessary for either gelatinization or reduction of "raw starch" flavor. For example, you can reduce the "raw" flavor in a water/flour slurry, or a beurre manie, by cooking the starch for a while. What the slurry and beirre manie does is coat the starch ganules so they they have greater dispersability in the liquid - to that they don't clump together and form a water-tight seal around clumps of granules (aka: lumps). One of the problems with the starch in four is cellulose that surrounds the starch granules. When it get's wet it get's sticky ... and it's not very water soluable.
Starch is a long chain of glucose molecules .... dextrinization is a breakdown of these long chains into shorter (usually) 3-5 molecule chains of sugars (dextrins). When you make a roux - the darker it gets the greater the dextrinization ... and inversly the lower the gelatinization ability since the starch chains have been reduced. Ironically, the color and flavor are probably due to a Maillard browning reaction (since we are dealing with proteins and sugars in the flour) ... and one of the characteristics of Maillard browning reactions and a colored roux is a "nutty" flavor.
As for the magic chemical reaction that causes a change in flavor ... it might be no more complex than allowing time for the starch granuals to hydrate and change from a straight chain into a helix configuration, under heat. This would change the way they bond to other molecules - thus altering flavor.
Another factor is the permeability of the starch granules. The cellulose capsule around the starch granules is not very water soluable ... but it is fat soluable ... thus in making a roux you "soften" up the cellulose and start the hydration process. Hydration is slow at first ... and with time and heat the rate increases.
Regarding your bechamel question ... do whatever works for you. I generally make mine in a double-boiler and don't whisk constantly. However, I have found that time and temp can make a difference texture - but little in flavor.
The hot roux to cold liquid, or hot liquid to cold roux, is a rule begging to be broken. It depends on what you are doing. When I make beef stews ... I don't flour the meat before searing, I don't add flour and essentially make a roux after adding the meat back to the pot .... when the stew is done I make a roux in a seperate saute pan and add the hot roux to the simmering stew ... and then cook it for about 3 minutes ... no muss, no fuss, no lumps (knock on wood).
Oh, regarding the milk proteins .... knew there was something I was missing. You can scourch them (mainly the milk solids which are more than just proteins) ... which will cause bitterness. Cooking them at a lower temp you can cook them just about all day without a problem.
As someone on another forum suggested, go to the library, or purchase, and read Shirley Corriher's Cookwise, and Harold McGee's On Food and Cooking (Revised Edition). I would also add to this the 1995 Scribner edition of Irma Rombauer's classic, Joy of Cooking.
I, like you, like to know how things work. But sometimes, it's possible to get so caught up in the "micro-molecular" workings that it becomes almost impossible to understand without a PhD in Chemistry - and doesn't alter the tried-n-proven techniques.
"It ain't what you don't know that gets you in trouble. It's what you know for sure that just ain't so." - Mark Twain