Of course, the main stay in Roux en Y GJ is to separate the bilio-pancreatic limb from the gastrojejunal limb so as to prevent reflux biliary gastritis. It's more demanding however effective. It can be made as classical roux en Y or modified one without any jejunal division by doing simple GJ followed by entero-enteric anastomosis between the afferent and efferent jejunal limbs to the stomach and then ligating the afferent limb DISTAL to the entero-enteric anastomosis. 

Whether simple loop gastro-jejunostomy or a Roux loop re-construction, this would usually be combined with a distal gastrectomy. This remains a perfectly acceptable intervention for distal gastric cancer and, of course, for many years thousands wre done as anti-ulcer operations before the real pathophysiology of peptic ulcer disease was understood.

In relation to the 2 options - a Roux loop clearly decrease the chance of bile reflux. In my experiance (well over 250 distal gastrectomies) although bile relux is very common it rarely causes suignificant sympotoms. If this does happen, a Roux loop reconstruction is relatively simple. The downside of using a Roux loop first up is the 'Roux syndrome' - not commonm but very difficult to treat. The Roux limb essentially becomes paretic and after trying all the various prokinetics patients not infrequently end up on jej tube feeding.You can probably tell where my prejudice lies1