Indeed Winflex uses an inflatable torus, but the blades are inside while they are outside for the rotary ballute kite which is a TRPT like Daisy. That said the attachment of the blade around the torus in the Winflex manner can be applied.
The torus could be the one (22 m in diameter (page 4), 220 m³ and 130 kg of lifting force (page 5)) on the video below:
The rigid blades are therefore fixed to the torus.
The first Daisy was an inflatable torus with a flat bottom, and with flexible kites around it (in the beginning of the video below):
That might be a path to follow.
And also: Investigation of Aerodynamic Characteristics of the Tilted Toroidal Ballute System | Aerospace Sciences Meetings , the abstract indicating a L/D ratio of 0.4 with a tilt angle of 40 degrees, at Mach 3.
I use a torus for several designs. This torus design could be more appliable than some others.
Nevertheless, for a large number of stacked units, the angular rotation speed of each of the rotors should be identical. One risky solution: placing increasingly larger rotors for increasingly stronger winds. Another solution: arranging to capture the angular speed of each rotor to equalize it for all rotors, preferably using methods less complex than variable pitch, possibly by means of telescopic blades or aerodynamic brakes.