Sorry, you are still missing a key reason for the existence of projects like WINE and DOSEMU.

Think proprietary business applications which talk to proprietary business hardware with binary-only drivers.

Think about running these applications and hardware on a free operating system.

Then think about using the facilities of that free operating system to reverse engineer the hardware interface from the driver.

Then think about implementing a free business application which talks to that free driver.

Suddenly the free software platform has increased in capability.

Or look at this scenario:

Proprietary application saves data in a proprietary data format.

Some curious hacker runs said application under WINE/DOSEMU and hooks the functions which write to the file.

Said hacker writes up a spec and sample programs to access that database.

Next release of OpenOffice supports opening that database natively.

Enabling this sort of development is the primary reason why these projects will not go away as long as there are people interested in promoting free platforms. People who want immediate practical benefits such as being able to play their binary-only games on a more stable and user-friendly system happen to get what they want too, as a side-effect of the more general push towards removing the chains of unsupported proprietary systems from users.