Still doesn't really work, it will give a very poor quality sinewave over a limited frequency range, useless for most purposes.
Have you at least simulated it?
Did you notice the dual ganged pot?
It's probably my fault for not using a pot symbols.
The triangle wave is pretty linear <1%.
The filter's cut-off frequency is over 3× the frequency of oscillation which gives a pretty decent sinewave, probably not as good as a Wein bridge or phase shift but certainly good enough for most applications. It also has the advantage of square and triangle outputs, although this isn't important here.
The THD of the waveform can be reduced further by reducing the amount of hysteresis on the Schmit trigger. Although this will reduce the amplitude of the waveform so will require more gain as well and increase the noise level.
This is the cheapest and crudest way of building a filtered Schmitt trigger oscillator, using a comparator plus dual op-amp might be better.
With the components shown, the frequency range is about 5kHz to 50kHz and the amplitude and quality of the sine wave doesn't change much at either extreme, so 100Hz to 600Hz should be trivial.
There's no reason why this circuit couldn't be used as the basis of a simple function generator, using a dual ganged rotary switch to change the capacitors, from 1nF to 100µF, it should be easy to go from 5Hz to 500kHz.
I'm not saying that this is the best way of doing it, all I'm saying is it shouldn't be ruled out yet.
Here's the circuit again, with the proper pot symbols and simulation results showing the sine wave with it set to about 15kHz.