A quest for long-lasting stretched linen surface

First some words about sizes. The purpose of a size is twofold. One is to cut or even out the absorbency of the substrate. The other is to isolate the substrate from the binder in the ground. Sizes were essential when using oil grounds on fabric to prevent the fabric from becoming greasy from the absorbed oil and to prevent the acids in the oil from attacking the cellulose in the canvas. Sizes were also applied to wooden panels, primarily to even out the absorbency of the substrate. The wood would have different absorbencies between the grain lines and the wood between them. Applying a gesso or even oil ground to an unsized panel may result in a surface that had alternating bands of more and less absorbency. This would be the death knell for a tradition egg tempera painting which relies on a evenly absorbent surface.
It is not necessary to apply a size to a fabric that will receive an acrylic dispersion ground. The binder in the ground would size the fabric as long as enough coats were applied. This is not to say that one should not size the fabric in such a case. Sizing with a couple of coats of acrylic dispersion medium or thinned acrylic dispersion gel creates a surface that better allows for a smooth application of acrylic dispersion ground. It also prevents the unsightly and uneven absorption of the ground to the reverse of the canvas. This is less disturbing to see on cotton duck but is really sloppy looking when the same is done on a linen support.
As long as ample applications of acrylic dispersion ground are applied to unsized fabric, it protects the canvas in the same manner as having sized it. I still maintain that the procedure I mentioned above creates a more satisfactory canvas.
I see no reason to use a PVA size below an acrylic dispersion ground. It provides no more rigidity (I have read that it provides less) and only adds another heterogenous material to the mix.  
As to your outline plan. I see no problem with sizing the linen with acrylic dispersion medium, applying a few coats of slightly thinned acrylic dispersion ground. The more coats, the more rigid the surface. The amount of water would depend on how viscous the ground is out of the jar. You do want some leveling before drying to retain the mechanical tooth. I would aim for ease of application but certainly not so thin that it is watery or able to penetrate through the size layers. I would apply these grounds with a wide spatula rather than a brush, or initially with a brush but smoothing with a wide spatula. It creates a far superior surface than does brush application alone unless one is intending on exploiting the inevitable textured, brushy surface. These layers would effectively fill in a good deal of the interstices  
After the above has dried a couple of days one could spatula on a coat of thinned lead white in linseed oil (you want a slurry that is runny but not watery . Apply an additional layer after it is allowed to dry enough to take a fingernail test and if absolutely necessary for the intended degree of smoothness.
Now after writing all of the above, many will tell you that the lead white ground is not absolutely necessary. I personally prefer the feel and the manner in which it accepts oil paint of a well applied lead white oil ground. Others disagree. I remember back when I was in art school, one of my classmates in a figure painting class who had always painted on self-prepared acrylic dispersion grounds on cotton duck canvases. One day they tries to work on a oil ground after seeing that I used them and our discussions about the difference. This student became so frustrated by the change is feel that they ended up slashing the canvas with an e-xacto knife before the end of that 6-hour studio class.
 I will ask our moderators from Golden Artists Colors to comment as well because they may have a different take on this. Additionally, they have done many, many tests on the rigidity of acrylic dispersion grounds and the adhesion/permanence of oil paints on such grounds