Objective To analyze the accuracy of the new generation intraocular lens (IOL) formula and the Wang-Koch (WK) adjustment in individuals with severe myopia undergoing combined anterior-posterior surgery. Methods A total of 100 patients (100 eyes) with axial length (AL) >26 mm undergoing combined anterior and posterior segment surgery in our department from June 2015 to June 2021 were enrolled in this study. Totally 13 IOL calculation formulas, such as Barrett Universal Ⅱ (BUⅡ), Emmetropia Verifying Optical (EVO), Kane, Haigis, Hoffer Q, Holladay, and SRK/T as well as the first generation linear WK eye axis optimization formula (Haigis-WK1, SRK/T-WK1, Hoffer Q-WK1, Holladay 1-WK1) and the second generation linear WK eye axis optimization formula (SRK/T-WK2, Holladay 1-WK2), were utilized to determine IOL diopter. The mean prediction error (ME), mean absolute error (MAE), median absolute error (MedAE), and percentage of prediction error within different refractive thresholds were calculated to assess the accuracy of each formula. In order to investigate the impact of AL on accuracy of IOL calculation, the patients were categorized into group A (38 eyes, 26.00 mm < AL≤28.00 mm), group B (28 eyes, 28.00 mm < AL≤30.00 mm), and group C (34 eyes, AL >30.00 mm). Additionally, the patients were also divided into silicone oil (63 eyes) and non-silicone oil (37 eyes) groups based on preoperative intraocular filling to determine the effect of intraocular filling on IOL calculation. Results The MedAE values of the new-generation IOL formulas (BUⅡ, EVO, and Kane) were significantly lower (0.34D, 0.31D, 0.35D) than those of traditional formulas (P < 0.05). Additionally, the percentage of prediction errors within ±0.25 D, ±0.50 D, ±0.75 D, and ±1.00 D were significantly higher with the new formulas when compared to traditional ones (P < 0.05). Traditional formulas exhibited a hyperopic deviation (0.35D~0.65 D), which could be corrected using the WK eye axis adjustment. WK2 was effective in improving the accuracy of calculation in SRK/T and Holladay 1 formulas, and there was no statistical difference in the MedAE values of SRK/T-WK2 and Holladay 1-WK2 formulas with the new-generation IOL formulas. When comparing the formulas across different axial eye length groups in terms of MedAE, the new IOL formulas (BUⅡ, kane and EVO) and WK-corrected formulas (Haigis-WK1, SRK/T-WK1, SRK/T-WK2, Hoffer Q-WK1, Holladay 1-WK1, Holladay 1-WK2) showed no significant differences among the different AL groups. However, there were significant differences in the MedAE values of traditional formulas among the different AL groups (P < 0.01), with group C displaying the largest MedAE values (Haigis: 0.84D, SRK/T: 1.10D, Hoffer Q: 1.23D, Holladay 1: 1.20D). Comparing the MedAE values of different formulas in various preoperative intraocular filling groups, it was observed that the MedAE values of the new-generation IOL formulas (BUⅡ, kane and EVO), Haigis, and SRK/T formulas were significantly higher in the silicone oil-filled eye group than the non-silicone oil-filled eye group (P < 0.05). Conclusion In patients with high myopia undergoing combined anterior and posterior segment surgery, the precision and consistency of the new-generation IOL formulas and the second-generation linear WK axis correction formula exhibit significant improvements when compared to traditional formulas.