Typically, companies will tell you that the incremental cost to continue drilling once you are out in the lateral is low, so if your leasehold can stand it, go ahead and keep drilling until the torque begins giving you problems.
The axiom is the longer the well, the more likely it is a better well.
In the past, a combination of tool reliability/costs and hole problems prevented longer laterals.
You still have to get that pipe to go down the hole (and come back out), so a lot depends on hole sidewall friction, tortuosity of the hole, and the formation itself. Yes, drilling fluid is important, too.
Tools (MWD, mud motors, and bits) have become considerably more reliable, and faster penetration rates have allowed more hole to be drilled before the wellbore becomes less stable. I think a lot depends on the formation you are drilling, too. The longest Bakken lateral I worked was on a 2 1/2 section spacing, with about 11,000 ft. of lateral. The last two thousand feet took as long as the first 5,000, but that was before rotary steerable assemblies and some of the current tools (in Elm Coulee Field, Richland Co., MT), and before the Bakken boom in North Dakota got rolling.