What you say is kind of being done with open-access repositories like arXiv (minus the requirement to include reference to the free version), and it has been a step forward. But publishing the same article in two journals is severely frowned upon in the scientific community. And even if this changed, journal publishers like Elsevier, Springer, etc. require authors to sign a copyright transfer agreement and forbid publication in any other journal. There is no way they are going to relinquish this.